This proposal connects the principal investigator's research on selective and divided attention with the development of an analysis of preparation in choice reaction time tasks. Support is requested for research on: (a) processing of and memory for unattended information, using variations on the selective attention reaction time (SART) procedure, (b) analysis of highly efficient timesharing of decisions, and (c) collection of data to guide theoretical analysis of preparation. The SART and timesharing procedures have been developed and used extensively in the principal investigator's prior work. The procedures used for the study of preparation are an outgrowth of recent findings of the timesharing research. A new development, for all three areas of research encompassed by this proposal, is the use of multiple regression techniques for analysis of reaction time data.